Selective production of CuSbS₂, Cu₃SbS₃, and Cu₃SbS₄ nanoparticles using a hot injection protocol

Abstract

Homogeneous Cu–Sb–S nanocrystals with several compositions can be synthesized in a solution through the hot-injection method. Photoelectrochemical analyses of films of these nanoparticles revealed that these nanoparticles have p-type semiconductive characters.

---

A ternary Cu–Sb–S system contains several compounds such as CuSbS₂ (chalcostibite),¹ Cu₁₂Sb₄S₁₃ (tetrahedrite),² Cu₃SbS₃ (skinnerite),³ and Cu₃SbS₄ (famatinite).^2,4 Owing to their differences in crystalline structures and atomic compositions, they would have unique optical and electric properties. Indeed, their measured and calculated optical band gaps (E_gs) are quite different. For instance, E_gs of CuSbS₂, Cu₃SbS₃ and Cu₃SbS₄ have been determined to be 1.56 eV, 1.89 eV, and 0.98 eV, respectively.⁵ These Cu–Sb–S compounds are regarded as promising sulfide materials for photovoltaic application: they are being considered as alternative absorbers to the chalcopyrite compound of Cu(In,Ga)(S,Se)₂ (CIGS) since they are composed of low-toxicity and abundant elements. Specifically, the CuSbS₂ compound has been studied widely because of its relatively optimal E_g for sun-light absorption, though there have been few reports showing appreciable solar cell properties⁶ due probably to the difficulty in obtaining a dense CuSbS₂ film without any shunts. Recently, we have successfully prepared a dense CuSbS₂ film by an electrochemical route; preliminary conversion efficiency of the solar cell based on the CuSbS₂ film reached 3.12%.⁷

It has been reported by several research groups that ternary and quaternary semiconductor nanoparticles, such as CIGS,⁸ CuInS₂,⁹ and Cu₂ZnSnS₄,¹⁰ were prepared via a precipitation reaction in hot organic solutions as precursors of dense thin film photoabsorbers as well as light-absorbing units without sintering them for sensitized solar cells. Hence, nanoparticles of Cu–Sb–S compounds having different crystalline phases would be promising starting units to fabricate solar cells composed of these new photoabsorbers. However, successful fabrication of Cu–Sb–S nanoparticles is limited to those with Cu₁₂Sb₄S₁₃ and Cu₃SbS₄ compositions.^2,4

The above-described literature studies motivated us to fabricate Cu–Sb–S nanoparticles having other compositions and crystalline phases. In this study, we achieved selective syntheses of CuSbS₂ and Cu₃SbS₃ nanoparticles for the first time. Selective formation of Cu₃SbS₄ nanoparticles was also demonstrated. Optical characteristics of these nanoparticles dispersed in a solution and photoelectrochemical (PEC) properties of them immobilized on an electrode substrate by using a layer-by-layer deposition technique were also investigated.

Cu–Sb–S nanoparticles were prepared through thermal reactions of metal chloride and elemental sulfur (S) or bis(trimethylsilyl) sulfide (TMS) in a hot solution composed of oleylamine (OA) and 1-octadecene (OD) under an Ar atmosphere. A schematic drawing of the procedures for synthesizing these nanoparticles is shown in Fig. 1. Experimental parameters are also summarized in Table S1.† Into an OA solution containing CuCl and SbCl₃, a certain amount of S or TMS dissolved in OA or OD was added. Then the mixture was heated for a certain duration. Kinds of sulfur sources, temperatures of the mixtures, and reaction durations varied depending on the target Cu–Sb–S nanoparticles. For the synthesis of CuSbS₂ nanoparticles, S dissolved in OA was used and the reaction was performed at 260 °C for 5 min. Cu₃SbS₄ nanoparticles were obtained by injection of S dissolved in OA into the OA solution containing CuCl and SbCl₃ at 200 °C; the reaction was continued at the same temperature for 10 min. Selective formation of Cu₃SbS₃ was achieved by mixing TMS dissolved in OD and the OA solution containing CuCl and SbCl₃ at 240 °C for 1 min followed by keeping the mixture at 150 °C for 30 min. After the reactions, large particles were removed from the resulting suspensions by centrifugation. Cu–Sb–S nanoparticles were isolated from the supernatant by precipitation with the addition of methanol.

Fig. 1 Flowchart for fabrication of Cu–Sb–S nanoparticles.

The XRD pattern of particles synthesized at 260 °C for 5 min using a 1 : 1 : 2.2 mixture of CuCl, SbCl₃ and S exhibited typical diffraction patterns of chalcostibite CuSbS₂ without any appreciable reflections of other compounds (Fig. S1a†). The XRD pattern of the sample obtained by heating a 1 : 1 : 4.4 mixture of CuCl, SbCl₃ and S at 200 °C for 10 min showed four broad reflections at 2θ of 28.5°, 32.5°, 47.5°, and 56.5°: these reflections are attributable to (112), (200), (204), and (312) reflections of famatinite Cu₃SbS₄, respectively (Fig. S1b†). A skinnerite Cu₃SbS₃ compound was also obtained by mixing TMS with CuCl and SbCl₃ with the Cu : Sb : S molecular ratio of 1 : 1 : 4.4 at 240 °C followed by keeping the mixture at 150 °C for 30 min (Fig. S1c†).

Fig. 2 shows TEM images of representative Cu–Sb–S nanoparticles. Monodispersed spherical particles were observed in the CuSbS₂ nanoparticles, whereas Cu₃SbS₄ and Cu₃SbS₃ samples exhibited polydispersed states having angular morphologies of particles. Based on corresponding size distribution plots of these samples obtained by measuring more than 200 particles (insets of Fig. 2a–c), average diameter (d_av) and standard deviation (σ) of these samples were determined; values of d_av(σ) were 14.3 nm (1.4 nm) for CuSbS₂, 10.5 nm (1.7 nm) for Cu₃SbS₄, and 10.8 nm (1.7 nm) for Cu₃SbS₃. Due to the requirement of a relatively high temperature during the synthesis, CuSbS₂ exhibits a relatively large d_av value. Reflecting polydispersed states, Cu₃SbS₄ and Cu₃SbS₃ samples have large σ values compared to that of the CuSbS₂ sample. As shown in Fig. 2d–f, high-resolution TEM images of these samples revealed that they had a single-crystalline nature: each particle had continuous lattice fringes throughout the particle.

Fig. 2 TEM images of (a and d) CuSbS₂, (b and e) Cu₃SbS₄ and (c and f) Cu₃SbS₃ nanoparticles.

Even by using the same 1 : 1 mixture of CuCl and SbCl₃ solution, we obtained three kinds of nanocrystals with different compositions. In order to examine mechanistic aspects of such selective formation of different nanocrystals, various samples obtained at different temperatures for several reaction durations were analyzed. For example, when the reaction was performed at temperatures less than 200 °C using a solution containing CuCl, SbCl₃ and S, a mixture of Cu₂S and Cu₃SbS₄ was initially formed; single-phase Cu₃SbS₄ was obtained by continuing the reaction at a given temperature. These results suggest that formation of Cu₃SbS₄ was through the reaction between initially formed Cu₂S nanocrystals and antimony species in the reaction solution. It should be noted that the formal charge of the antimony component in the Cu₃SbS₄ crystalline is pentavalent (Sb(V)) despite the use of a trivalent antimony (Sb(III)) source. Since the reaction was performed in an Ar atmosphere, oxidation of the Sb(III) component during the formation of Cu₃SbS₄ was likely to be induced by remaining or contaminated oxygen species in the reaction solution.

In separate experiments using mixtures of CuCl/S and SbCl₃/S, formation of Cu₂S was confirmed to occur primarily at a relatively low temperature, whereas a binary antimony sulfide (Sb₂S₃) was formed at temperatures higher than 240 °C. Moreover, when the synthesis of Cu–Sb–S nanoparticles was performed at 240 °C, a mixture of Cu₂S, Sb₂S₃, CuSbS₂ and Cu₃SbS₃ was obtained. These results indicate that formation of CuSbS₂ and Cu₃SbS₃ nanocrystals occurs through the reaction of Cu₂S and Sb₂S₃. This is the reason for the requirement of a relatively high temperature for the formation of single-phase CuSbS₂. For the synthesis of Cu₃SbS₃, it should be necessary to avoid formation of Sb(V) because the formal charge of antimony in Cu₃SbS₃ was trivalent. Thus, Sb(III) species in the solution was initially reacted to form Sb₂S₃ at a relatively high temperature in a short period. Then the growth of nanocrystalline Cu₃SbS₃ should be continued at a relatively low temperature (≤150 °C) in order to avoid formation of CuSbS₂ as well as Cu₃SbS₄. Moreover, for selective synthesis of the Cu₃SbS₃ phase, use of TMS was more suitable than S as a sulfur source: use of S induced contamination of appreciable amounts of Cu₃SbS₄ phase. Relatively high reactivity of TMS with impurity water to produce S²⁻ (or H₂S) would lead to the facilitation of Sb₂S₃ formation with inhibiting oxidation of Sb(III) into Sb(V).

Semi-quantitative analyses of Cu, Sb and S contents of prepared nanoparticles by EDX indicated that compositions of these nanoparticles were almost stoichiometric with the exception of the Cu₃SbS₃ sample that was obviously Cu-poor (Cu/Sb/S = 35/17/48) compared to the stoichiometric one (Cu/Sn/Zn = 43/14/43). The above-discussed formation mechanism of the sample suggests that this deviation is due to the insufficient formation of Cu₂S, leading to remaining appreciable amounts of Sb₂S₃ in the final nanoparticles.

Absorption spectra of Cu–Sb–S nanoparticles uniformly suspended in toluene are given in Fig. S2.† The Cu₃SbS₄ sample showed the absorption onset at ca. 1330 nm, which is similar to the reported spectrum of Cu₃SbS₄ nanoparticles (Fig. S2b†).^2,4 CuSbS₂ and Cu₃SbS₃ samples showed absorption onsets at relatively shorter wavelength regions, as shown in Fig. S2a and c.† It should be noted that all of the spectra have appreciable baseline components, suggesting the presence of tiny amounts of contaminants having relatively narrow band gap energies such as Cu₂S and Cu₃SbS₄ (for CuSbS₂ and Cu₃SbS₃), while they were not confirmed by the above XRD analyses. From intersects of the linear portions of the curves with the wavelength axis, E_gs of CuSbS₂, Cu₃SbS₄ and Cu₃SbS₃ nanoparticles were estimated to be 1.53 eV, 1.72 eV and 0.93 eV, respectively. Deviations from reported E_gs (see above) are likely to be attributed to inaccuracy in the determination of intersects owing to the presence of baseline components.

In order to determine PEC properties of the Cu–Sb–S nanoparticles, thus-obtained nanoparticles were immobilized on an ITO-coated glass (ITO/glass) substrate. The process for fabrication of these electrodes is described in ESI.† Current density–voltage characteristics of particle films were investigated by linear sweep voltammetry (LSV) in an aqueous Eu(III) solution at pH adjusted to 4. Fig. 3 shows typical LSV plots of the Cu–Sb–S particle films. Since cathodic photocurrents appeared, all of the films prepared in the present study behaved as p-type semiconductor photoelectrodes. As determined by using the lock-in technique, photocurrent onsets of these samples were in the range of −0.1 V to 0.1 V (Fig. S3†).¹¹ Although accurate analyses could not be performed because of their weak photoresponses, these results suggest that they have similar energy positions of their valence band edges (E_VBs). Due to the appreciable differences of E_gs, conduction band edges (E_CBs) of these nanoparticles should be variable. For the application of these p-type compounds to solar cells, therefore, it is required to find suitable n-type compounds to form p–n heterojunctions in view of energy offsets (ΔE_CBs) between E_CBs of p-type and n-type compounds.¹¹ In order to study ΔE_CBs more accurately, construction of bulk films having sufficient photoresponses by using these nanoparticles as starting materials is now in progress.

Fig. 3 LSV curves of (a) CuSbS₂, (b) Cu₃SbS₄ and (c) Cu₃SbS₃ particulate films deposited on an ITO/glass substrate.

Conclusions

CuSbS₂, Cu₃SbS₃ and Cu₃SbS₄ nanoparticles with p-type semiconductive properties were successfully synthesized through thermal reactions of metallic ions and S or TMS in high-temperature OA or OA/OD mixed solution. Reaction temperatures higher than 260 °C gave a pure CuSbS₂ crystal phase, whereas the Cu₃SbS₄ compound was selectively formed when the reaction temperatures were fixed less than 200 °C. For the synthesis of Cu₃SbS₃ nanoparticle, a two-step heating profile, including generations of both Cu₂S and Sb₂S₃ compounds at 150 °C for a short period followed by reaction between these binary sulfides at 240 °C, was found to be required. We can expect to construct novel light energy conversion systems using these nanoparticles when combined with appropriate n-type compounds.

Acknowledgements

This work was carried out as part of a program supported by NEDO Japan and a Grant-in-Aid for Scientific Research on Innovative Areas (All Nippon Artificial Photosynthesis Project for Living Earth) from MEXT Japan. Financial support by The Murata Science Foundation is acknowledged. Dr Takao Sakata (Osaka University) is also acknowledged for their help in TEM measurements.

Notes and references

1. H. Su, Y. Xie, S. Wan, B. Li and Y. Qian, Solid State Ionics, 1999, 123, 319; Y. Rodriguez-Lazcano, M. T. S. Nair and P. K. Nair, J. Cryst. Growth, 2001, 223, 399; A. Rabhi, M. Kanzari and B. Rezig, Mater. Lett., 2008, 62, 3576; C. Garza, S. Shaji, A. Arato, E. P. Tijerina, G. A. Castillo, T. K. Das Roy and B. Krishnan, Sol. Energy Mater. Sol. Cells, 2011, 95, 2001; J. T. R. Dufton, A. Walsh, P. M. Panchmatia, L. M. Peter, D. Colombara and M. Saiful Islam, Phys. Chem. Chem. Phys., 2012, 14, 7229; D. J. Temple, A. B. Kehoe, J. P. Allen, G. W. Watson and D. O. Scanlon, J. Phys. Chem. C, 2012, 116, 7334.
2. J. van Embden, K. Latham, N. W. Duffy and Y. Tachibana, J. Am. Chem. Soc., 2013, 135, 11562.
3. D. Chen, G. Shen, K. Tang, X. Jiang, L. Huang, Y. Jin and Y. Qian, Mater. Res. Bull., 2003, 38, 509; M. X. Wang, G. H. Yue and P. X. Yan, J. Cryst. Growth, 2008, 310, 3062.
4. J. van Embden and Y. Tachibana, J. Mater. Chem., 2012, 22, 11466.
5. L. Yu, R. S. Kokenyesi, D. A. Keszler and A. Zunger, Adv. Energy Mater., 2013, 3, 43–48.
6. Y. Rodriguez-Lazcano, M. T. S. Nair and P. K. Nair, J. Electrochem. Soc., 2005, 152, G635.
7. W. Septina, S. Ikeda, Y. Iga, T. Harada and M. Matsumura, Thin Solid Films, 2014, 550, 700.
8. J. Tang, S. Hinds, S. O. Kelley and E. H. Sargent, Chem. Mater., 2008, 20, 6906; Q. J. Guo, G. M. Ford, R. Agrawal and H. W. Hillhouse, Prog. Photovolt: Res. Appl., 2013, 21, 64.
9. S. L. Castro, S. G. Bailey, R. P. Raffaelle, K. K. Banger and A. F. Hepp, J. Phys. Chem. B, 2004, 108, 12429; M. G. Panthani, V. Akhavan, B. Goodfellow, J. P. Schmidtke, L. Dunn, A. Dodabalapur, P. F. Barbara and B. A. Korgel, J. Am. Chem. Soc., 2008, 130, 16770.
10. Q. J. Guo, H. W. Hillhouse and R. Agrawal, J. Am. Chem. Soc., 2009, 131, 11672; S. C. Riha, B. A. Parkinson and A. L. Prieto, J. Am. Chem. Soc., 2009, 131, 12054; C. Steinhagen, M. G. Panthani, V. Akhavan, B. Goodfellow, B. Koo and B. A. Korgel, J. Am. Chem. Soc., 2009, 131, 12554; T. Kameyama, T. Osaki, K.-i. Okazaki, T. Shibayama, A. Kudo, S. Kuwabatade and T. Torimoto, J. Mater. Chem., 2010, 20, 5319; A. Khare, A. W. Wills, L. M. Ammerman, D. J. Norrisz and E. S. Aydil, Chem. Commun., 2011, 47, 11721; M. Li, W.-H. Zhou, J. Guo, Y.-L. Zhou, Z.-L. Hou, J. Jiao, Z.-J. Zhou, Z.-L. Du and S.-X. Wu, J. Phys. Chem. C, 2012, 116, 26507; G. M. Ford, Q. Guo, R. Agrawal and H. W. Hillhouse, Chem. Mater., 2011, 23, 2626.
11. R. Klenk, Thin solid films, 2001, 387, 135; M. Turcu and U. Rau, Thin Solid Films, 2003, 432, 158.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and additional figures. See DOI: 10.1039/c4ra07648f

---